[lld/mac] Implement -sectalign

[thakis]

Summary

clang sometimes passes this flag along (see D68351), so we should implement it.

[int3]

lgtm

[thakis]

Thanks! Landing with a slightly stronger test (but same code).

lld/MachO/Config.h

 enum class UndefinedSymbolTreatment {
   unknown,
   error,
   warning,
   suppress,
   dynamic_lookup,
 };
 
+struct SectionAlign {
+  llvm::StringRef segName;
+  llvm::StringRef sectName;
+  uint32_t align;
+};
+
 struct SegmentProtection {
   llvm::StringRef name;
   uint32_t maxProt;
   uint32_t initProt;
 };
 
 class SymbolPatterns {
 public:
UndefinedSymbolTreatment undefinedSymbolTreatment =
       UndefinedSymbolTreatment::error;
   llvm::MachO::HeaderFileType outputType;
   std::vector<llvm::StringRef> systemLibraryRoots;
   std::vector<llvm::StringRef> librarySearchPaths;
   std::vector<llvm::StringRef> frameworkSearchPaths;
   std::vector<llvm::StringRef> runtimePaths;
   std::vector<std::string> astPaths;
   std::vector<Symbol *> explicitUndefineds;
-  // There are typically very few custom segmentProtections, so use a vector
-  // instead of a map.
+  // There are typically few custom sectionAlignments or segmentProtections,
+  // so use a vector instead of a map.
+  std::vector<SectionAlign> sectionAlignments;
   std::vector<SegmentProtection> segmentProtections;
 
   llvm::DenseMap<llvm::StringRef, SymbolPriorityEntry> priorities;
   SectionRenameMap sectionRenameMap;
   SegmentRenameMap segmentRenameMap;
 
   SymbolPatterns exportedSymbols;
   SymbolPatterns unexportedSymbols;

lld/MachO/Driver.cpp

for (char c : protStr) {
     default:
       error("unknown -segprot letter '" + Twine(c) + "' in " + protStr);
       return 0;
     }
   }
   return prot;
 }
 
+static std::vector<SectionAlign> parseSectAlign(const opt::InputArgList &args) {
+  std::vector<SectionAlign> sectAligns;
+  for (const Arg *arg : args.filtered(OPT_sectalign)) {
+    StringRef segName = arg->getValue(0);
+    StringRef sectName = arg->getValue(1);
+    StringRef alignStr = arg->getValue(2);
+    if (alignStr.startswith("0x") || alignStr.startswith("0X"))
+      alignStr = alignStr.drop_front(2);
+    uint32_t align;
+    if (alignStr.getAsInteger(16, align)) {
+      error("-sectalign: failed to parse '" + StringRef(arg->getValue(2)) +
+            "' as number");
+      continue;
+    }
+    if (!isPowerOf2_32(align)) {
+      error("-sectalign: '" + StringRef(arg->getValue(2)) +
+            "' (in base 16) not a power of two");
+      continue;
+    }
+    sectAligns.push_back({segName, sectName, align});
+  }
+  return sectAligns;
+}
+
 static bool dataConstDefault(const InputArgList &args) {
   switch (config->outputType) {
   case MH_EXECUTE:
     return !args.hasArg(OPT_no_pie);
   case MH_BUNDLE:
     // FIXME: return false when -final_name ...
     // has prefix "/System/Library/UserEventPlugins/"
     // or matches "/usr/libexec/locationd" "/usr/libexec/terminusd"
config->sectionRenameMap[{validName(arg->getValue(0)),
                               validName(arg->getValue(1))}] = {
         validName(arg->getValue(2)), validName(arg->getValue(3))};
   }
   for (const Arg *arg : args.filtered(OPT_rename_segment)) {
     config->segmentRenameMap[validName(arg->getValue(0))] =
         validName(arg->getValue(1));
   }
 
+  config->sectionAlignments = parseSectAlign(args);
+
   for (const Arg *arg : args.filtered(OPT_segprot)) {
     StringRef segName = arg->getValue(0);
     uint32_t maxProt = parseProtection(arg->getValue(1));
     uint32_t initProt = parseProtection(arg->getValue(2));
     if (maxProt != initProt && config->arch() != AK_i386)
       error("invalid argument '" + arg->getAsString(args) +
             "': max and init must be the same for non-i386 archs");
     if (segName == segment_names::linkEdit)

lld/MachO/Options.td

 def segalign : Separate<["-"], "segalign">,
     MetaVarName<"<boundary>">,
     HelpText<"Align all segments to hex power-of-2 <boundary>">,
     Flags<[HelpHidden]>,
     Group<grp_rare>;
 def sectalign : MultiArg<["-"], "sectalign", 3>,
     MetaVarName<"<segment> <section> <boundary>">,
     HelpText<"Align <section> within <segment> to hex power-of-2 <boundary>">,
-    Flags<[HelpHidden]>,
     Group<grp_rare>;
 def stack_addr : Separate<["-"], "stack_addr">,
     MetaVarName<"<address>">,
     HelpText<"Initialize stack pointer to hex <address> rounded to a page boundary">,
     Flags<[HelpHidden]>,
     Group<grp_rare>;
 def segprot : MultiArg<["-"], "segprot", 3>,
     MetaVarName<"<segment> <max> <init>">,

lld/MachO/OutputSegment.cpp

size_t OutputSegment::numNonHiddenSections() const {
   for (const OutputSection *osec : sections)
     count += (!osec->isHidden() ? 1 : 0);
   return count;
 }
 
 void OutputSegment::addOutputSection(OutputSection *osec) {
   osec->parent = this;
   sections.push_back(osec);
+
+  for (const SectionAlign &sectAlign : config->sectionAlignments)
+    if (sectAlign.segName == name && sectAlign.sectName == osec->name)
+      osec->align = sectAlign.align;
 }
 
 static DenseMap<StringRef, OutputSegment *> nameToOutputSegment;
 std::vector<OutputSegment *> macho::outputSegments;
 
 OutputSegment *macho::getOrCreateOutputSegment(StringRef name) {
   OutputSegment *&segRef = nameToOutputSegment[name];
   if (segRef)

lld/test/MachO/sectalign.s

+# REQUIRES: x86
+# RUN: llvm-mc -filetype=obj -triple=x86_64-apple-darwin %s -o %t.o
+
+# RUN: not %lld -dylib -o %t %t.o -sectalign __TEXT __text asdf 2>&1 \
+# RUN:     | FileCheck --check-prefix=NONUM -DNUM=asdf %s
+# RUN: not %lld -dylib -o %t %t.o -sectalign __TEXT __text 0x0X4 2>&1 \
+# RUN:     | FileCheck --check-prefix=NONUM -DNUM=0x0X4 %s
+# NONUM: error: -sectalign: failed to parse '[[NUM]]' as number
+
+# RUN: not %lld -dylib -o %t %t.o -sectalign __TEXT __text 16 2>&1 \
+# RUN:     | FileCheck --check-prefix=NOPOW -DNUM=16 %s
+# RUN: not %lld -dylib -o %t %t.o -sectalign __TEXT __text 0x16 2>&1 \
+# RUN:     | FileCheck --check-prefix=NOPOW -DNUM=0x16 %s
+# RUN: not %lld -dylib -o %t %t.o -sectalign __TEXT __text 0 2>&1 \
+# RUN:     | FileCheck --check-prefix=NOPOW -DNUM=0 %s
+# NOPOW:  error: -sectalign: '[[NUM]]' (in base 16) not a power of two
+
+## __DATA_CONST tests that the flag applies to names after section renaming.
+# RUN: %lld -dylib -o %t %t.o -sectalign __TEXT __text 20 \
+# RUN:                        -sectalign __DATA_CONST __const 0x40
+# RUN: llvm-readobj --section-headers %t \
+# RUN:     | FileCheck -DSECT=__text -DSEG=__TEXT -DALIGN=5 %s
+# RUN: llvm-readobj --section-headers %t \
+# RUN:     | FileCheck -DSECT=__const -DSEG=__DATA_CONST -DALIGN=6 %s
+
+# RUN: %lld -dylib -o %t %t.o -rename_section __TEXT __text __TxT __foo \
+# RUN:                        -sectalign __TxT __foo 0x40
+# RUN: llvm-readobj --section-headers %t \
+# RUN:     | FileCheck -DSECT=__foo -DSEG=__TxT -DALIGN=6 %s
+
+# CHECK:      Name: [[SECT]]
+# CHECK-NEXT: Segment: [[SEG]]
+# CHECK-NEXT: Address:
+# CHECK-NEXT: Size:
+# CHECK-NEXT: Offset:
+# CHECK-NEXT: Alignment: [[ALIGN]]
+
+.section __TEXT,__text
+.space 1
+
+.section __DATA,__const
+.space 1